The statistical legacy of P.C. Mahalanobis
He was instrumental in orchestrating different aspects of planning in the newly-independent India
In 2015, Barack Obama appointed D.J. Patil as the US’ first chief data scientist and deputy chief technology officer for data policy. Is it so important to get insights from loads of data for framing national policy in today’s world?
On 29 June, the nation celebrated the 125th birth anniversary of Prof. Prasanta Chandra Mahalanobis, the founder of the Indian Statistical Institute (ISI) and the creator of an enriched statistical legacy in India. At this juncture, my quest is to analyse how a Mahalanobis style of academic leadership and data handling is still the best way for the country’s benefit.
Mahalanobis established the ISI in 1931 when the subject of statistics was in its infancy. Elsewhere in the world, around that time, Jerzy Neyman founded the Statistical Laboratory at the University of California, Berkeley, in 1938, and their statistics department in 1955; Beijing University started its statistics department in 1956. In addition to ISI, Mahalanobis was also instrumental in establishing the statistics department at Calcutta University in 1941, the first full-fledged department in Asia offering a post-graduate degree in statistics. Mahalanobis, undoubtedly, was far ahead of his time.
Mahalanobis organized the first statistics conference in India in 1938 in Kolkata, with R.A. Fisher, perhaps the best statistician of the world at that time, as its president. As early as in 1933, he established Sankhyā: The Indian Journal Of Statistics, a milestone in the history of science in India. Three decades on, Mahalanobis would consider this “an adventure, even foolhardy”. However, sometimes such adventures do change the course of history.
Mahalanobis was a professor of physics, who became a statistician by chance. However, according to Prof. C.R. Rao, perhaps the most well-known statistician in the world today, Mahalanobis was a statistician by instinct and an economist by conviction. His statistical learning and the process of “increasing the efficiency of human effort” were mostly developed on the wings of data, built on economics and anthropology, primarily.
It all started in 1917, when Sir Brajendra Nath Seal requested the 24-year-old Mahalanobis to analyse the exam records of Calcutta University. Three years later, N. Annandale, the then director of Zoological and Anthropological Survey of India, gave a dataset of about 300 Anglo-Indians of Calcutta to Mahalanobis, which was instrumental in initiating his more than a decade-long research on anthropometric data, resulting in the theorizing of the “Mahalanobis distance”, his most notable research contribution.
Certainly, the volume of data Mahalanobis had to handle in those days was really small in today’s context. Some datasets, however, were really big. In the 1920s, he analysed rainfall and flood data of Odisha and Bengal of 50-60 years, and obtained very interesting, counterintuitive and useful insights, which led to the construction of the Hirakud Dam and the Durgapur Barrage, subsequently. At the early stage of the statistical culture, Mahalanobis carried out a number of large-scale sample surveys; he implemented pilot surveys, optimal surveys and interpenetrating subsamples in his applications. In the 1930s, he conducted an important survey on Bengal’s jute; it was observed that the results of his survey to find the total acreage under jute crop were even more accurate than complete enumeration. Latter-day ISI students even called him the “Professor of Counting and Measurement”, using the initials of his name.
Mahalanobis was instrumental in orchestrating different aspects of planning in the newly-independent India, including the second five-year plan. His extensive experience in carrying out surveys helped in establishing the National Sample Survey Office (NSSO).
The NSSO carried out some very useful surveys to gauge the poverty and inequality of the newly-independent nation, immensely helping policymakers. Angus Dayton, the 2015 economics Nobel laureate, used NSSO data in some of his influential research. Mahalanobis handled different kinds of data when statistics, the subject, was creeping, and computational help was also very limited. To facilitate bigger computations, the ISI procured India’s first computer in 1956 and the second in 1959.
Besides, Mahalanobis had the knack of spotting talent, mostly due to his extraordinary academic leadership.
We are now obsessed with Big Data, and aspire to use it in every aspect of our life—from business to sports. However, when it comes to the delicate question of national policy making, it becomes really serious, as a wrong decision based on potentially inaccurate or insufficient data analytics might have a tremendous impact on the nation.
One needs to be a top-level statistician with immense experience of handling data at the same time to be able to take the right policy decision. Mahalanobis’ adventure of paving the way of statistical heritage in India was supplemented by the fact that he himself was a world-class statistician. Unfortunately, even four and half decades after his demise, I cannot think of a person in this country having more expertise than him for framing data-based policy decisions for human welfare and national development—he certainly had a better understanding of data than most people in the business do.
Atanu Biswas is professor of statistics at the Indian Statistical Institute, Kolkata.
Comments are welcome at email@example.com
Editor's Picks »
- India to put former top climate change official Rajendra Pachauri on trial for sexual harassment
- Rahul Gandhi hits out at KCR, claims Telangana reeling under debt
- Deve Gowda-Siddaramaiah display rare bonhomie ahead of Karnataka by-polls
- Govt allocates Rs 144 crore to AYUSH ministry for alternative medicines
- Esperanto, A language whose time never came
- Policy rethink and higher volumes to aid container shippers
- DCB Bank delivers a strong Q2 but pressure on margins foreseen
- Havells India: Rising costs give a jolt to profitability in September quarter
- All’s well at Mindtree, except for high client concentration risk
- India’s rising steel demand is making companies starry-eyed